Datasets:
ai-forever
commited on
Commit
•
af5d23c
1
Parent(s):
0cd43f2
Update README.md
Browse files
README.md
CHANGED
@@ -217,11 +217,13 @@ An example of "test" looks as follows
|
|
217 |
|
218 |
#### Annotation process
|
219 |
|
220 |
-
|
|
|
|
|
221 |
|
222 |
#### Who are the annotators?
|
223 |
|
224 |
-
|
225 |
|
226 |
### Personal and Sensitive Information
|
227 |
|
|
|
217 |
|
218 |
#### Annotation process
|
219 |
|
220 |
+
We set up two-stage annotation project via a crowd-sourcing platform Toloka:
|
221 |
+
1. Data gathering stage: we provide the texts with possible mistakes to annotators and ask them to write the sentence correctly;
|
222 |
+
2. Validation stage: we provide annotators with the pair of sentences (source and its corresponding correction from the previous stage) and ask them to check if the correction is right.
|
223 |
|
224 |
#### Who are the annotators?
|
225 |
|
226 |
+
Native Russian speakers who passed the language exam.
|
227 |
|
228 |
### Personal and Sensitive Information
|
229 |
|